INSAIT is proud to announce its participation in this year’s NeurIPS conference in Vancouver, taking place from December 10-15, 2024. Visit us at Booth No. 47 to learn more about our groundbreaking work and explore exciting opportunities in AI research and development.
We are actively inviting applications for tenure-track and tenured faculty, postdoctoral researchers, and PhD students in all areas of AI. Our world-class facilities, competitive packages, and collaborative environment offer exceptional opportunities for advancing your career while contributing to state-of-the-art research in AI.
Research Highlights
INSAIT will showcase its latest projects in foundational models for robotics, multimodal learning, 3D vision, transfer learning, and mathematical reasoning using large language models. Several of these have been accepted at NeurIPS 2024 in the main conference and workshops.
Key papers from INSAIT include:
Sharing Key Semantics in Transformer Makes Efficient Image Restoration
Poster Session 3, Poster #1007, East Exhibit Hall A-C
Thursday, December 12, 11 a.m. – 2 p.m. PST
DAGER: Exact Gradient Inversion for Large Language Models
Poster Session 3, Poster #6201, West Ballroom A-D
Thursday, December 12, 11 a.m. – 2 p.m. PST
Implicit Zoo: A Large-Scale Dataset of Neural Implicit Functions for 2D Images and 3D Scenes
Datasets and Benchmarks, Poster #5402, West Ballroom A-D
Thursday, December 12, 11 a.m. – 2 p.m. PST
SPEAR: Exact Gradient Inversion of Batches in Federated Learning
Poster Session 4, Poster #6301, West Ballroom A-D
Thursday, December 12, 4:30 p.m. – 7:30 p.m. PST
Constraint-Based Synthetic Data Generation for LLM Mathematical Reasoning
MATH-AI: The 4th Workshop on Mathematical Reasoning and AI
West Meeting Room 118-120
Saturday, December 14
Learning Generative Interactive Environments by Trained Agent Exploration
D3S3 Workshop: Data-driven and Differentiable Simulations, Surrogates, and Solvers
West Meeting Room 116-117
Sunday, December 15, 11 a.m. – 12 p.m. PST
Incentivizing Truthful Collaboration in Heterogeneous Federated Learning
Optimization for ML Workshop
West Ballroom A
Sunday, December 15, 3 p.m. – 4 p.m. PST
Meet the Team
This year, INSAIT’s contributions include work authored by Prof. Luc Van Gool, Prof. Martin Vechev, Ivo Petrov, Dimitar Dimitrov, Qi Ma, Dr. Danda Pani Paudel, Nikita Tsoy, Kristian Minchev, Dr. Nikola Konstantinov, Nedko Savov, and INSAIT summer interns Timofey Fedoseev, Naser Kazemi, and Dimitar Chakarov.
Join Us
We look forward to connecting with attendees, exchanging ideas, and exploring potential collaborations. Let’s discuss how you can contribute to groundbreaking advancements in AI research. Visit us at Booth No. 47!